Direct photon and neutral pion production in pp and Pb-Pb collisions measured with the ALICE experiment at LHC
NNovember 5, 2018 22:32 WSPC Proceedings - 9.75in x 6.5in Peressounko page 1 Direct photon and neutral pion production in pp and Pb–Pb collisionsmeasured with the ALICE experiment at LHC
D. Peressounko ∗ for the ALICE Collaboration RRC ”Kurchatov institute”,Kurchatov sq.,1, Moscow, 123182, Russia ∗ E-mail: [email protected]
Measurements of direct photon and neutral pion production in heavy-ion collisionsprovide a comprehensive set of observables characterizing properties of the hot QCDmedium. Direct photons provide means to test the initial stage of an AA collision andcarry information about the temperature and space-time evolution of the hot medium.Neutral pion suppression probes the parton energy loss in the hot medium. Measure-ments of neutral meson spectra in pp collisions at LHC energies √ s =0.9, 2.76, 7 TeVserve as a reference for heavy-ion collisions and also provide valuable input data for pa-rameterization of the QCD parton Fragmentation Functions. In this talk, results fromthe ALICE experiment on direct photon and neutral pion production in pp and Pb–Pbcollisions are summarized. Keywords : Direct photons; neutral pions; quark-gluon matter.
1. Introduction
The ultimate goal of heavy ion collisions is a detailed study of the properties ofquark-gluon matter. Photons provide tools for testing almost all key features of thishot matter: spectra and correlations of hard hadrons, which can be reconstructedusing photon decays, test the energy loss of energetic partons in hot matter andthe collective flow of identified hadrons up to high p T . Direct photons, definedas photons not from hadronic decays, are made up of prompt photons (emittedby partons of colliding nucleons) and thermal photons (emitted by the hot matteranalogously to blackbody radiation). The former allows us to probe the inital stateof the collision while the latter reflects the temperature and the space-time evolutionof the matter. In addition, the thermal-photon flow reflects the development of thehot matter’s collective flow at all stages of the collision.
2. Experimental Setup
A detailed description of the ALICE experimental setup can be found in . Com-pared to other LHC experiments, ALICE excels at low- p T physics and advancedparticle identification. The core of ALICE is the central tracking system consistingof the Inner Tracking System and the Time Projection Chamber. Charged particleidentification is improved by the Transition Radiation Detector and Time Of Flightdetector. Finally, ALICE has two calorimeters, the PHoton Spectrometer (PHOS) a r X i v : . [ nu c l - e x ] D ec ovember 5, 2018 22:32 WSPC Proceedings - 9.75in x 6.5in Peressounko page 2 and the electromagetic calorimeter (EMCal), as well as a set of smaller detectorsfor triggering and characterizing events. D a t a / F i t NN s ) c (GeV/ T p D a t a / F i t NN s
20 40% Pb Pb NN s ) c (GeV/ T p NN s
40 60% Pb Pb NN s
10 20% Pb Pb ) c (GeV/ T p NN s
60 80% Pb Pb
Fig. 1. Comparison of π spectra measured in Pb–Pb collisions at √ s NN = 2 .
76 TeV in 6centrality bins with PHOS and PCM techniques. Both spectra are divided by the fit to thecombined spectrum .
3. Neutral Pion Measurements
Photons can be reconstructed in ALICE in several ways: using traditional calorime-try with the PHOS and EMCal or by the Photon Conversion Method (PCM) viareconstructing e + e − tracks from photons conversion in the central tracking system .PHOS has fine granularity leading to excellent energy and position resolution thoughit has a relatively small acceptance. PCM provides good position and energy resolu-tion and full 2 π coverage in the azimuth. However, since ALICE was constructed tominimize the material budget, the photon conversion probability before the middleof the TPC, where tracks still can be reconstructed with high efficiency, is about8%. As a result, both methods have comparable acceptance × efficiencies.The ability to simultaneously measure photon and neutral pion spectra withseveral independent detectors improves the reliability of the final results. Moreover,the PCM and PHOS measurements have distinct systematic uncertainties, oppositedependencies of p T resolution and different sensitivities to pileup. The agreementbetween the two measurements is shown in fig. 1, where the π spectra in differentcentrality bins in Pb–Pb collisions at √ s NN = 2 .
76 TeV are compared. To elucidatethe comparison both spectra are divided by the fit to the combined spectrum.Within the statistical and systematic uncertainties the measurements are in goodagreement . ovember 5, 2018 22:32 WSPC Proceedings - 9.75in x 6.5in Peressounko page 3 combined Spec. T p = 0.5 µ NLO T p = µ NLO T p = 2 µ NLO (BKK) T p = 2 µ NLO = 7 TeV (*) s , π syst., stat. = 2.76 TeV (**) s , π syst., stat. = 0.9 TeV (*) s , π syst., stat. (*) PLB 717 (2012) 162 172 (**) arXiv:1405.3794 f i t N L O = 7 TeV s , π f i t N L O = 2.76 TeV s , π ) c (GeV/ T p f i t N L O = 0.9 TeV s , π ALI−DER−82346
Fig. 2. Comparison of π spectra measured in pp collisions at √ s = 0 . , .
76 and 7 TeV withseveral pQCD predictions. For clarity both data and predictions are divided by the fit to data.
We compare the π spectrum in pp collisions at √ s = 0 . , .
76 and 7 TeV withseveral selected pQCD predictions in fig. 2. To make the comparison of steeplyfalling spectra clear, we fit our spectra to a function and plot ratios of the measuredspectra to the fit along with the pQCD predictions to the fit . pQCD approximatelyreproduces the hadron yield at the lowest LHC energy ( √ s = 0 . √ s = 2 . , . In contrast, pQCD well reproduces jet spectra inpp collisions at LHC energies . A possible explanation is that gluon fragmentationbecomes increasingly important, while the gluon fragmentation functions are notwell restricted by existing data at lower √ s . This uncertainty should be reducedin the recent calculation of fragmentation functions , which includes these ALICEresults in their global analysis.ALICE measured neutral pion production in Pb–Pb collisionsat √ s NN = 2 .
76 TeV with two techniques: PHOS and PCM. Combined spectrawere produced for 6 centrality classes. For a quantitative estimate of the parton ovember 5, 2018 22:32 WSPC Proceedings - 9.75in x 6.5in Peressounko page 4 energy loss, the nuclear modification factor R AA ( p T ) = d N/ d p T d y | AA (cid:104) T AA (cid:105) × d σ/ d p T d y | pp (1)was calculated, where the nuclear overlap function (cid:104) T AA (cid:105) is related to the aver-age number of inelastic nucleon-nucleon collisions (cid:104) N coll (cid:105) and the pp inelastic crosssection σ ppinel as (cid:104) T AA (cid:105) = (cid:104) N coll (cid:105) /σ ppinel . The neutral pion nuclear modification factorsfor 6 centrality bins are shown in fig. 3. At p T (cid:38) c pions from hard interac-tions dominate and the strong suppression in R AA reflects considerable energy lossfor hard partons. At low p T ( (cid:46) c ) the spectrum is defined by collective (hy-drodynamic) expansion and comparing with pp is not particularly meaningful. Theintermediate p T region transitions between these two regimes. In the most centralcollisions R AA reaches a minimum of ∼ . . Since the spectra of initialpartons at LHC energies is considerably harder, stronger suppression means muchlarger energy loss compared to RHIC.Fig. 3 compares the measured nuclear modification factors with predictions fromtwo advanced models. Currently, there is no consensus as to which features are mostimportant and should be incorporated in descriptions of energy loss in AA collisions.In the Vitev et al. model , in addition to collisional energy loss, initial state effectsare included, while Horowitz et al. account for geometrical fluctuations of hardprocesses. Basically, both models are close to data, though Horowitz’s is not asgood in reproducing the centrality dependence. AA R NN s π GLV WHDG ) c (GeV/ T p AA R NN s
20 40% Pb Pb ALICE π NN s π ) c (GeV/ T p NN s
40 60% Pb Pb ALICE π NN s
10 20% Pb Pb ALICE π ) c (GeV/ T p π = 2.76 TeV NN s
60 80% Pb Pb
Fig. 3. Neutral pion nuclear modification factor in Pb–Pb collisions at √ s NN = 2 .
76 GeV in6 centrality classes . For comparison, predictions of models of Vitev (GLV) and Horowitz(WHDG) are shown. ovember 5, 2018 22:32 WSPC Proceedings - 9.75in x 6.5in Peressounko page 5
4. Direct photon spectra and flow
ALICE measures the direct photon yield via a statistical approach; the estimateddecay photon spectrum is subtracted from the measured inclusive photon spectrum.In this analysis one first constructs a double ratio R γ = γ inclusive /π measured γ decay /π param ≈ γ inclusive γ decay . (2)A double ratio of unity, R γ = 1, represents no direct photon yield whereasan increase above unity constitutes a presence of direct photons. The advantageof this approach is that most of the largest systematic uncertainties cancel in theratio. Preliminary ALICE results are shown in fig. 4(a) for central and fig. 4(b) forperipheral events. Blue bands show a pQCD prediction of the contribution fromprompt direct photons . ALICE finds no excess of direct photons in peripheralevents as data agree with pQCD predictions within the uncertainties. In centralcollisions, the direct photon yield agrees with the prompt photon yield at high p T but indicate additional excess at low p T . (GeV/c) T p ) π / de c a y γ ) / ( π / i n c γ ( Direct photon double ratio ) decay γ / direct,pp,NLO γ coll NLO prediction: 1 + (N T = 0.5,1.0,2.0 p µ for = 2.76 TeV NN s0 40% Pb Pb, ALI−PREL−27956 (a) (GeV/c) T p ) π / de c a y γ ) / ( π / i n c γ ( Direct photon double ratio ) decay γ / direct,pp,NLO γ coll NLO prediction: 1 + (N T = 0.5,1.0,2.0 p µ for = 2.76 TeV NN s40 80% Pb Pb, ALI−PREL−28000 (b)Fig. 4. Photon double ratios in central (a) and peripheral (b) Pb–Pb collisions at √ s NN = 2 .
76 TeV.
From the measured double ratio (fig. 4), the direct photon spectrum is derivedas N dirγ = (1 − /R γ ) N inclusiveγ and plotted in fig. 5. For comparison, pQCD pre-dictions are shown by the blue band. In addition, in the region p T < c where one expects a dominant contribution of thermal direct photons, ALICE fitthe data and extracted the inverse slope. However, one should keep in mind thatthis slope is influenced by collective expansion and the entire evolution of the systemand therefore should not be directly interpreted as the temperature in the center ofthe fireball.The collective flow of direct photons is extremely interesting because it is ex-pected that direct photons are emitted from the hottest stage of the collision and ovember 5, 2018 22:32 WSPC Proceedings - 9.75in x 6.5in Peressounko page 6 (GeV/c) T p ) c ( G e V d y T dp T p N d e v . N π = 2.76 TeV NN s0 40% Pb Pb, Direct photons (scaled pp) T = 0.5,1.0,2.0 p µ Direct photon NLO for 51 MeV ± /T), T = 304 T exp( p × Exponential fit: A
ALI−PREL−27968
Fig. 5. Direct photon spectrum in central (0-40%) Pb–Pb collisions at √ s NN = 2 .
76 TeV. Theblue line represents pQCD prompt photon predictions while the red line is an exponential fit inthe range 0 . < p T < . c . their flow reflects the development of collective expansion at early stages. Experi-mentally, one can measure the flow of inclusive photons, estimate the flow of decayphotons and estimate the flow of direct photons as v γ,dirn = v γ,incln R γ − v γ,decn − R γ , (3)where v γ,dirn , v γ,incln and v γ,decn are the flow of direct, inclusive and decay photonswith respect to the n th harmonic. As one can see from equation 3, with R γ close tounity, the uncertainties rapidly increase. Therefore, we present the results of directphoton flow as a comparison of collective flow of inclusive and decay photons, seefigs. 6 and 7 for elliptic and triangular flow respectively. In fig. 6a (7a) the elliptic(triangular) flow contribution from inclusive photons is shown in red while the decayphotons are shown in black. Predictions are also shown for Next-to-Leading-Order(NLO) prompt photons plus decay photons (blue dash-dotted line) as well as twomodels for inclusive photons (decay + prompt + thermal) in the red and green curves. Quantitative comparisons are shown in the right insets of these figures aswell. The differences are shown in units of the sigma of the total uncertainties. Wefound that both elliptic and triangular flow of inclusive photons are considerablysmaller than the expected flow of decay photons and agree with predictions of bothmodels incorporating thermal photon contributions. ovember 5, 2018 22:32 WSPC Proceedings - 9.75in x 6.5in Peressounko page 7 ) c (GeV/ T p γ v ,decay γ v ,incl γ v ALICE preliminary = 2.76 TeV NN s ALI−PREL−75753 (a) t o t. σ ) / , de c a y γ v , i n c l γ v (
505 ALICE decay cocktail = 2.76 TeV NN s t o t. σ ) / ,t heo . γ v , i n c l γ v (
505 ALICE decay cocktail + NLOPhys.Rev. D50 (1994) 1901 19160 1 2 3 4 5 6 t o t. σ ) / ,t heo . γ v , i n c l γ v (
505 ALICE decay cocktail + NLO+ thermal (Shen et al.)arXiv:1308.2111 ) c (GeV/ T p t o t. σ ) / ,t heo . γ v , i n c l γ v (
505 ALICE decay cocktail + NLO+ thermal (Holopainen et al.)Phys.Rev. C84 (2011) 064903
ALI−PREL−75765 (b)Fig. 6. (a) Comparison of elliptic flow of inclusive and decay photons. Lines represent contribu-tions of decay photons with theoretical calculations . (b) Difference between inclusive anddecay (top plot) and inclusive and theory (3 bottom plots) elliptic flows in units of total error. ) c (GeV/ T p γ v ,decay γ v ,incl γ v ALICE preliminary = 2.76 TeV NN s ALI−PREL−75777 (a) t o t. σ ) / , de c a y γ v , i n c l γ v (
505 ALICE decay cocktail = 2.76 TeV NN s t o t. σ ) / ,t heo . γ v , i n c l γ v (
505 ALICE decay cocktail + NLOPhys.Rev. D50 (1994) 1901 19160 1 2 3 4 5 6 t o t. σ ) / ,t heo . γ v , i n c l γ v (
505 ALICE decay cocktail + NLO+ thermal (Shen et al.)arXiv:1308.2111 ) c (GeV/ T p t o t. σ ) / ,t heo . γ v , i n c l γ v (
505 ALICE decay cocktail + NLO+ thermal (Chatterjee et al.)arXiv:1401.7464
ALI−PREL−75789 (b)Fig. 7. Same as fig. 6 but for triangular flow.
5. Conclusions
We reviewed the ALICE results on production of neutral pions in pp collisions at √ s = 0 .
9, 2.76 and 7 TeV and neutral pions and direct photons in Pb–Pb collisionsat √ s NN = 2 .
76 TeV. QCD calculations reproduce the neutral pion spectra in ppcollisions at √ s = 0 . ovember 5, 2018 22:32 WSPC Proceedings - 9.75in x 6.5in Peressounko page 8 direct photon spectrum agrees with pQCD predictions at p T > c but showsexcess at lower p T . Collective flow of inclusive photons differs from estimates ofcollective flow of decay photons and agrees with expectations including contributionsfrom thermal photons. Acknowledgments
This work was partially supported by the grant RFBR 12-02-91527.
References
1. K. Aamodt et al. , The ALICE experiment at the CERN LHC,
JINST , p.S08002 (2008).2. B. B. Abelev et al. , Neutral pion production at midrapidity in pp and Pb-Pbcollisions at √ s NN = 2.76 TeV, Eur.Phys.J.
C74 , p. 3108 (2014).3. B. Abelev et al. , Neutral pion and η meson production in proton-proton colli-sions at √ s = 0 . √ s = 7 TeV, Phys.Lett.
B717 , 162 (2012).4. D. d’Enterria, K. J. Eskola, I. Helenius and H. Paukkunen, LHC data challengesthe contemporary parton-to-hadron fragmentation functions,
PoS
DIS2014 , p.148 (2014).5. G. Aad et al. , Measurement of inclusive jet and dijet production in pp collisionsat √ s = 7 TeV using the ATLAS detector, Phys.Rev.
D86 , p. 014022 (2012).6. D. de Florian, R. Sassot, M. Epele, R. J. Hernandez-Pinto and M. Stratmann,Parton-to-Pion Fragmentation Reloaded, arXiv:1410.6027 (2014).7. S. Adler et al. , A Detailed Study of High-p(T) Neutral Pion Suppression andAzimuthal Anisotropy in Au+Au Collisions at √ s NN = 200 GeV, Phys.Rev.
C76 , p. 034904 (2007).8. R. Sharma, I. Vitev and B.-W. Zhang, Light-cone wave function approach toopen heavy flavor dynamics in QCD matter,
Phys.Rev.
C80 , p. 054902 (2009).9. R. Neufeld, I. Vitev and B.-W. Zhang, A possible determination of the quarkradiation length in cold nuclear matter,
Phys.Lett.
B704 , 590 (2011).10. W. Horowitz and M. Gyulassy, The Surprising Transparency of the sQGP atLHC,
Nucl.Phys.
A872 , 265 (2011).11. L. Gordon and W. Vogelsang, Polarized and unpolarized isolated prompt pho-ton production beyond the leading order,
Phys.Rev.
D50 , 1901 (1994).12. W. Vogelsang and M. R. Whalley, A compilation of data on single and doubleprompt photon production in hadron hadron interactions,
J. Phys.
G23 , A1(1997).13. W. Vogelsang. Private communications (August, 2004).14. C. Shen, U. W. Heinz, J.-F. Paquet, I. Kozlov and C. Gale, Anisotropic flow ofthermal photons as a quark-gluon plasma viscometer, arXiv:1308.2111 (2013).15. H. Holopainen, S. Rasanen and K. J. Eskola, Elliptic flow of thermal photonsin heavy-ion collisions at Relativistic Heavy Ion Collider and Large HadronCollider,
Phys.Rev.